VoLUT: Efficient Volumetric Streaming Enhanced by LUT-based Super-Resolution

Chendong Wang, Anlan Zhang, Yifan Yang, Lili Qiu, Feng Qian, Suman Banerjee

Conference on Machine Learning and Systems 2025 · Day 4 · Session 12: Edge and Cloud Systems

The talk introduces **VoLUT**, a groundbreaking system designed to address the significant challenges of streaming high-fidelity volumetric video. Volumetric video, which offers a full 3D representation of a scene allowing for six degrees of freedom (6DOF) movement, promises truly immersive experiences for applications ranging from real-time telepresence to next-generation gaming and AR/VR. However, its inherently data-heavy nature far exceeds the capabilities of most consumer internet connections, making real-time streaming impractical. VoLUT tackles this by proposing a novel approach that leverages **lookup tables (LUTs)** for 3D super-resolution (SR), enabling efficient and high-quality volumetric video delivery even on resource-constrained embedded devices.

AI review

VoLUT is a legitimately interesting systems paper — replacing neural inference with LUT lookups for 3D super-resolution is a clever engineering trade-off, and the embedded hardware results are the kind of thing that makes a practical difference. But the write-up reads like an abstract padded into a talk summary, and the gap between 'this is interesting research' and 'here's how you build it' is wide enough that I couldn't reproduce the core quantization scheme or LUT construction from what's described. Solid MLSys work, but presented at a level of abstraction that limits its usefulness to…